The biggest obstacle to human-like artificial intelligence is inside your head

We live in a world where computers can trade stocks faster than any broker on the floor and best any living human in the game of Jeopardy.

But it turns out even our smartest artificial intelligence (AI) is still relatively dumb.

So what do humans have that our machines don’t?

For one, it’s our seemingly common-sense ability to take what we know and apply it to new situations. Imagine talking to someone in a loud bar. Though you can’t hear everything they’re saying, you can plug in the gaps based on the context of the conversation and what you know about the person to understand what they’re trying to say.

According to Bruno Olshausen, head of the Redwood Center for Theoretical Neuroscience at the University of California Berkeley, computers don’t work like that like that — they can’t fill in the gaps on the spot like a human can.

“Memory is central to cognition,” Olshausen told Bloomberg Business.

The human brain doesn’t completely remember everything that happens in a day. Instead, it summarises the day and pulls out highlights when the details become relevant again, Olshausen told Bloomberg.

But that’s not the only problem stopping scientists from building human-like AI: No one understands how our brains work to create intelligence.

To that end, the National Institutes of Health is devoting $US300 million to the Brain Initiative, a project to find and explore mechanisms that allow brains to store and retrieve information. Even the project website admits that “the human brain remains one of the greatest mysteries in science and one of the greatest challenges in medicine.”

Even now scientists are still trying to figure out how thousands of neurons work together to produce a physical action, like typing or reaching for a glass, according to the New York Times.

We also don’t understand consciousness, sentience, and “self-hood,” and we have no way to right now, according to computer scientist Stuart Russell.

“There is no scientific theory that could lead us from a detailed map of every single neuron in someone’s brain to telling us how that physical system would generate a conscious experience,” Russell told Tech Insider. “It’s not just that: We could not know how the brain works, in the sense that we don’t know how the brain produces intelligence.”

Olshausen says neuroscience today is akin to where physics was in the 1600s — we’ve got a long way to go before we can figure out human intelligence, much less emulate it in code.

“Where we are in terms of our understanding of the brain — and what it takes to build an intelligent system — is kind of pre-Newton,” Olshausen told Bloomberg. “If you’re in physics and pre-Newtonian, you’re not even close to building a rocket ship.”

NOW WATCH: Scientists are having robots play ‘Minecraft’ to learn about human logic

NOW WATCH: Briefing videos

Business Insider Emails & Alerts

Site highlights each day to your inbox.

Follow Business Insider Australia on Facebook, Twitter, LinkedIn, and Instagram.